[bjsneto]

Summary

Fixes EncloseColumnName returning already-delimited input without escaping embedded backticks.

The method now treats the input as an unquoted column name, escapes backticks inside it, and adds the enclosing backticks. QuoteSingle and QuoteDouble continue to add delimiters without escaping their input, preserving the existing Escape().QuoteSingle() production flow and avoiding double-escaping.

Regression tests were added for embedded and already-delimited backticks, as well as the quote helper behavior.

The existing bulk-copy regression test was corrected to pass each column name as a separate ColumnNames element instead of passing two comma-separated identifiers as one value.

Fixes #316

Checklist

• Unit tests covering the common scenarios were added
• A human-readable description of the changes was provided to include in CHANGELOG

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[CLAassistant]

All committers have signed the CLA.

[cursor]

Cursor Bugbot has reviewed your changes using high effort and found 1 potential issue.

^{Reviewed by Cursor Bugbot for commit 5d6792e. Configure here.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[polyglotAI-bot]

❌ Blocker: Document the breaking column-name contract

This changes the public InsertBinaryAsync(..., columns, ...) and ClickHouseBulkCopy.ColumnNames contract even though the helper is internal. An element containing the SQL spelling `id` previously selected column id; it now selects a name containing literal backticks. Existing callers therefore fail, or target a different column if both names exist. docs/http.mdx has no update, and the changelog fragment only describes the internal helper. Document the raw-name requirement in the bulk-insert section, show callers replacing previously quoted names with raw names, and record the compatibility change in the changelog. Also state that each collection element is one name.

_{A reply here is read on the next review: an explanation that holds up drops the finding.}

[polyglotAI-bot]

Summary
The HTTP client now treats column names as raw names and always adds delimiters in the quote helpers. The identifier fix is incomplete for backslash/backtick combinations, and the breaking change to insert column-name handling is not documented. Block pending these changes.

❌ Blockers

• ClickHouse.Driver/Utility/StringExtensions.cs:38 Escape existing backslashes before backticks so a raw name cannot terminate the generated identifier.
• ClickHouse.Driver/Utility/StringExtensions.cs:29 Document the changed interpretation of previously quoted column names in the public insert APIs and provide migration guidance.

Tests

• Add a server round-trip for a column whose raw name starts and ends with backticks and contains a backslash immediately before an interior backtick. Exercise the default schema probe, UseSchemaCache, and explicit ColumnTypes, and assert the inserted value.

Missing context

• Review was static; this sandbox does not permit builds or server tests. A server round-trip is needed to validate the escaping through the public insert path.

Final verdict

❌ Block. Complete backslash escaping and document the public column-name contract change.

_{Automated review by Polyglot AI (gpt-6-astra) of d01d8ba. Requested by @alex-clickhouse. Replies on the inline threads are read on the next review. A maintainer can ask for a new review by mentioning @polyglotAI-bot or requesting a review from it.}

[polyglotAI-bot]

Review of d01d8ba posted: see the summary comment and the inline threads.

[Copilot]

🔵 Needs a closer look

Distinguishing tests for already-delimited QuoteSingle and QuoteDouble inputs are still needed.

0 open findings

🧠 Review effort: Lite